Sampling Size

S = (z/e)^2

90% = z= 1.65
E = 5% = 0.05
(1.65/0.05)^2 = 1089

Given :-
95% = z= 1.96
E =?
S=169
Formula:-
S = (z/e)^2
Solution:-
169 = (1.96/e)^2
13=1.96/e
E=1.96/13 = 0.150
E= 0.150 x 100 = 15%
ANOVA
Car – Dealer – 20 mileages

10 people scross

H0 = Null = 20
H1 =Alternate =/20

Upper – 22
Lower – 18
Accept h1 reject h0, dealer is liar.

H0 = All are same

H1 = Anyone of them is different
T Paired Test:
To observe changes in output. Basically, before after and pre post difference.
Two tail Hypothesis :
H0 : There is no difference meaning d = 0
H1: There is a difference meaning d # 0
One tail Hypothesis:
H0: There is no difference meaning d = 0
H1: There is an increase in mean d>0
OR
H0: There is no difference meaning d = 0
H1: There is a decrease in mean d<0
Interpretation:
T stat > critical value = reject H0, accept H1
T stat < critical value = reject H1, accept H0

Chi square test/ test for association:

H0: There is no association
H1: There is an association
Interpretation:
If calculated x2> tabulated x2 = accept H1 ; if calculated x2< tabulated x2 = accept H0
If given p value > 0.05 = accept H0 ; if given p value < 0.05 = accept H1

ANOVA:
One way ANOVA:

H0: All are same

H1: Any one of them is different.

Two-way ANOVA (Salesmen and areas)

For salesmen:

H0: All salesmen perform the same.

H1: At least one salesman performs differently.

For Areas:

H0: All areas have same effect.

H1: At least one area is different.

For both one way and two-way Anova :

If calculated f > tabulated f = accept h1 ; if calculated f < tabulated f = accept h0

1a. Correlation vs. Chi-Square

Correlation and chi-square are both statistical tools used to analyze relationships between
variables, but they differ significantly in their application, purpose, and types of data they
handle.

Correlation is a statistical method used to measure the strength and direction of a

relationship between two continuous variables. The most commonly used correlation
technique is Pearson’s correlation coefficient, which ranges from -1 to +1. A positive
correlation indicates that as one variable increases, the other also increases. A negative
correlation suggests that as one variable increases, the other decreases. A correlation of zero
means there is no relationship between the two variables.

For example, if researchers study the relationship between hours spent studying and exam
performance, they can use correlation analysis to determine whether more study time leads to
higher test scores.

On the other hand, Chi-square (χ²) analysis is a test used for categorical data. It helps to
determine whether there is a significant association between two categorical variables—often
used in surveys and research studies involving nominal (label-based) variables. The chi-
square test compares observed frequencies with expected frequencies to check whether they
differ significantly due to random chance or an actual relationship.

For instance, if a researcher examines whether gender influences food preferences (e.g., male
vs. female vs. non-binary individuals choosing vegetarian or non-vegetarian meals), the chi-
square test would evaluate whether the observed choices are statistically different from
expected probabilities.

Key Differences Between Correlation and Chi-Square:

 Type of Data: Correlation is used for continuous numerical data, while chi-square is
used for categorical data.
 Purpose: Correlation measures the strength of relationships between variables, while
chi-square assesses whether observed differences are statistically significant.
 Interpretation: Correlation coefficients provide information on the direction and
magnitude of a relationship, while chi-square provides significance but does not
indicate strength or direction.
Both methods are widely used in research, business analytics, and social sciences, providing
valuable insights depending on the nature of the data.

1b. Benefits and Limitations of Primary Data

Primary data refers to data that is collected firsthand by a researcher for a specific purpose. It
is original and not derived from existing sources, which makes it highly relevant and reliable
for specific studies. However, it comes with its own advantages and challenges.

Benefits of Primary Data:

1. Firsthand Information: Since primary data is collected directly by the researcher, it

is fresh, accurate, and tailored to the research objectives.
2. Specific to Needs: Unlike secondary data, which is pre-collected for other purposes,
primary data is customized based on the exact needs of the study.
3. Greater Control: Researchers have full control over how the data is gathered,
ensuring ethical and unbiased practices.
4. Up-to-Date: Primary data is collected in real-time, making it more relevant to current
conditions.
5. Ensures Reliability: Because primary data is collected directly from the source, it is
often more reliable compared to older secondary data.

Limitations of Primary Data:

1. Time-Consuming: Gathering primary data requires careful planning, execution, and

analysis, which takes more time than accessing pre-existing secondary data.
2. Expensive: Conducting surveys, interviews, experiments, or observations can be
costly, requiring substantial resources and manpower.
3. Potential Biases: If the data collection process lacks proper methodology, it could
lead to biases in results, affecting the accuracy of the findings.
4. Limited Historical Data: Primary data only reflects current findings, making it
difficult to analyze long-term trends without past data.
5. Requires Expertise: Designing effective research instruments and conducting valid
analysis often require expert knowledge.

Despite these challenges, primary data is invaluable for research studies, market analysis,
academic projects, and decision-making processes in various fields.
4a. Focus Group Discussion Procedure

A focus group discussion (FGD) is a qualitative research technique used to gather insights
from a group of individuals on a particular topic. It allows researchers to explore perceptions,
opinions, and ideas in a conversational setting. Here’s a step-by-step guide to conducting an
effective focus group discussion:

Step 1: Define Objectives and Select Participants

The first step is to clarify the purpose of the discussion. Researchers must define what they
aim to achieve, whether it's understanding consumer preferences, testing a new product, or
exploring social attitudes. Participants are selected based on relevant criteria such as
demographics, experience, or expertise in the subject matter. Ideally, a group consists of 6–12
people to maintain a manageable discussion.

Step 2: Prepare Questions and Topics

A well-structured set of questions guides the discussion. Questions should be open-ended,

encouraging participants to share their thoughts and experiences freely. The discussion
typically follows a structured format with topics progressing logically from general to
specific.

Step 3: Conduct the Discussion with a Facilitator

A moderator or facilitator leads the session, ensuring all participants have an opportunity to
speak. The facilitator’s role is crucial in managing the discussion flow, keeping participants
engaged, and preventing dominance by a single individual. The session usually lasts 1–2
hours, allowing enough time for in-depth discussion while maintaining focus.

Step 4: Record Responses and Analyse Findings

Discussions are recorded via audio, video, or notes for later analysis. Researchers identify
recurring themes, opinions, and insights from the conversation, summarizing key findings
that can be used for decision-making or further research.
Focus group discussions provide rich qualitative data, making them a valuable tool for
businesses, social research, product development, and policy-making.

5b. The research process consists of several key steps that help ensure a study is systematic,
thorough, and credible. Here’s a breakdown of the different stages involved:

1. Identifying the Research Problem

The first step in any research process is defining the problem or topic that needs
investigation. Researchers must ensure the problem is relevant, specific, and feasible for
study.

2. Reviewing Existing Literature

Before conducting original research, scholars review previous studies, books, and articles on
the topic to understand existing knowledge, identify gaps, and refine their own approach.

3. Formulating Research Objectives and Hypothesis

Based on the research problem, clear objectives are established. If applicable, a hypothesis (a
testable statement predicting the outcome) is formulated to guide the study.

4. Selecting Research Design and Methodology

Researchers decide on the type of study—whether qualitative or quantitative—and choose

appropriate methods such as experiments, surveys, case studies, or observations.

5. Data Collection

This stage involves gathering relevant data through primary methods (e.g., interviews,
experiments, or surveys) or secondary sources (e.g., books, official reports, or online
databases).

6. Data Analysis and Interpretation

Once collected, data is analysed using statistical tools, qualitative assessments, or

computational techniques. Results are then interpreted to draw meaningful conclusions.
7. Drawing Conclusions and Making Recommendations

Based on findings, researchers summarize key takeaways, validate or refute their hypothesis,
and offer practical recommendations if applicable.

8. Reporting and Publishing the Research

Finally, the research is documented in reports, academic papers, presentations, or

publications, making the findings accessible to relevant audiences.

7B. High-quality research is distinguished by several key features that ensure reliability,
accuracy, and meaningful contributions to knowledge. Here are the most important
characteristics:

1. Clarity and Well-Defined Objectives

The research should have clearly stated objectives and purpose. A well-defined problem
statement guides the study, ensuring focus and relevance.

2. Rigor and Systematic Approach

A structured methodology with logical steps ensures the study follows scientific principles,
minimizing errors and biases.

3. Ethical Integrity

Research should adhere to ethical standards, ensuring honesty, transparency, and respect for
participants, data privacy, and intellectual property rights.

4. Validity and Reliability

High-quality research ensures that findings are valid (accurately measuring what is intended)
and reliable (consistent across different studies or conditions).

5. Comprehensive Literature Review

A thorough review of existing knowledge helps establish context, identify gaps, and avoid
repetition of previous studies.

6. Proper Data Collection and Analysis

The methods for gathering and analysing data should be appropriate for the research
objectives, ensuring accuracy and meaningful interpretation.

7. Objectivity and Neutrality

Research should be free from personal biases or external influences, presenting findings
based on facts rather than opinions.

8. Clear and Logical Presentation of Results

Findings should be communicated in an organized, well-explained manner through reports,

articles, or presentations, making them accessible and useful to others.

9. Contribution to Existing Knowledge

The study should provide new insights, challenge existing theories, or offer practical
applications to help advance knowledge in a particular field.

10. Replicability and Generalization

A well-conducted research study should be replicable by other researchers and applicable to

similar situations beyond the immediate study sample.
8a. Snowball sampling and purposive sampling are both non-probability sampling
techniques, meaning they do not rely on random selection. However, they differ in their
approach and purpose.

1. Snowball Sampling

Snowball sampling is used when it is difficult to identify or access members of a population

directly. Researchers start with a small number of known participants who then refer or
recruit others from their network, creating a "snowball effect."

Key Features:

 Commonly used in hard-to-reach populations (e.g., drug users, underground artists,

refugees).
 Relies on participant referrals to expand the sample.
 Risk of bias, as networks may lead to a homogeneous sample.

Example:

If a researcher wants to study undocumented immigrants, they may begin with a few known
individuals who then introduce others with similar experiences.

2. Purposive Sampling

Purposive sampling involves deliberately selecting participants based on specific criteria

relevant to the study. Researchers choose individuals who have particular characteristics,
experiences, or expertise related to the research question.

Key Features:

 Used when researchers want insightful, targeted data rather than general population
representation.

 Participants are selected based on predefined characteristics rather than referrals.

 Helps ensure diverse, expert, or specific perspectives are included.

Example:

If a study examines the impact of leadership styles in organizations, researchers may

intentionally select experienced managers or CEOs rather than a random sample of
employees.
Key difference:-

Snowball Sampling Purposive Sampling

Aspect

Selection Process Uses referrals to build sample Researchers handpick participants

Target
Hard-to-reach or hidden groups Individuals with specific characteristics
Population

High (network-based
Bias Risk Lower, but depends on selection criteria
recruitment)

Social research, hidden Case studies, expert opinions, qualitative

Usage Scenario
communities research